Rituximab therapy modulates IFN-γ and IL-4 gene expression in a patient with acquired haemophilia A


Several published observations have demonstrated that acquired haemophilia A can be successfully treated with rituximab (Franchini, 2007). Interestingly, emerging reports demonstrated that T cell-mediated diseases, such as ulcerative colitis and psoriasis, are sometimes exacerbated by the use of rituximab (Dass et al, 2007; Goetz et al, 2007).

Currently, the impact of anti-B-cell therapy on the immune system is not clearly understood. T-helper (Th) cells are divided into Th1, Th2 and Th17 subtypes producing γ-interferon (IFN-γ), interleukin (IL)-4 and IL-17, respectively. Alterations affecting the balance between Th cell types need to be monitored as these may cause the emergence of pathologies (Lucey et al, 1996). Here, we investigate the impact of rituximab on the gene-expression of key T-helper cytokines: IFNG, IL4 and IL17A.

This is the case of an 18-year-old woman, previously healthy G1P2 (delivered twins 18 months earlier) who presented with excessive bleeding after appendectomy. On work-up, she was found to have a prolonged activated partial thromboplastin time (aPTT), normal prothrombin time (PT), and factor VIII (FVIII) of 3·5%; a FVIII inhibitor was measured at 7 Bethesda units (BU). Treatment with high doses (60 units/kg per h of FVIII concentrate via continuous infusion provided good haemostasis with average FVIII levels of 83% during a complicated 2 month hospital course; however, the inhibitor persisted during this period. A decision was made to attempt inhibitor eradication with rituximab.

The study was approved by the Tulane University Institutional Review Board and written informed consent for participation was obtained from the patient. Peripheral blood mononuclear cells (PBMC) were isolated by Ficoll-Paque™Plus (GE Healthcare Bio-Sciences AB, Uppsala, Sweden) gradient separation from heparinized blood. Cells were surface-stained with the following fluorochrome-conjugated anti-human antibodies: CD3-peridinin chlorophyll protein Cyanin 5·5 (PerCp-Cy5·5) (clone SK7; BD Biosciences, San Jose, CA, USA), CD4-AmCyan (clone SK3; BD Biosciences), CD14-phycoerythrin-texas red (ECD) (clone RMO52; Beckman coulter, Miami, FL, USA), CD19-allophycocyanin (APC) (clone HIB19; BD Biosciences). Data acquisitions were performed using a BD LSRII Cytometer (BD Biosciences) and results were analysed with DiVa Software (BD Biosciences).

Expression of IFNG, IL4 and IL17A mRNA was monitored using reverse transcription polymerase chain reaction (RT-PCR). Total RNA was isolated from whole blood using the PaxGene blood RNA isolation kit (BD-Qiagen, Germantown, MD, USA) according to the manufacturer’s instructions. RNA quality was assessed using Experion Automated Electrophoresis Station (Bio-Rad, Hercules, CA, USA). The reverse transcription reaction was performed on 1 μg total RNA using iScript™ cDNA synthesis Kit (Bio-Rad). RT-PCR was performed on a Bio-Rad iCycler with iQ™ Supermix (BioRad) according to manufacturer’s instructions. The PCR programme cycle was the following: 50°C 2 min, 95°C 3 min, 40 cycles of 95°C 15 s 60°C 1 min. The following TaqMan® probes from Applied Biosystems (Foster City, CA, USA) were used: Hs99999903_m1 (ACTB); Hs00174143_m1 (IFNG); Hs00929862_m1 (IL4) and Hs00936345_m1 (IL17A). Cycle thresholds were normalized internally to ACTB. The proportion of transcript present in the samples was calculated using the 2 ΔΔ CT calculation method. The final results represent the relative amount of amplicon in samples (fold increase) with the level of transcripts collected at the onset of the treatment set to 1.

The FVIII inhibitor became undetectable 4 days after the first dose of rituximab (375 mg/m2) and FVIII levels initially increased as the FVIII inhibitor decreased (Fig 1). The FVIII level then gradually decreased again, while the FVIII inhibitor rose slightly. The patient therefore received subsequent doses of rituximab on days 26, 36 and 48. With these infusions, the FVIII level increased gradually to 112% on day 110 and remained in the normal range (116%) on day 235 (Fig 1).

Figure 1.

 Effect of B cell depletion on FVIII and FVIII inhibitor levels: Arrows indicate time points where rituximab doses (375 mg/m2 per dose) were administrated to the patient.

As expected, the initial dose of rituximab profoundly depleted CD19+ B-cells in the blood (data not shown). CD19+ B cells remained undetectable at day 110 but significant repopulation of PBMC by CD19+ cells were observed at day 235 (data not shown).

Furthermore, the first dose of rituximab was associated with a 4-fold increase of IFNG mRNA level (P < 0·01; Fig 2), and a 12-fold decrease in IL4 mRNA expression (P < 0·01). After peaking at day 26, IFNG mRNA level decreased to its nadir at day 48. Between day 26 and the end of the study, IL4 mRNA level rose steadily and recovered by day 110. At day 235, the expression of IL4 mRNA returned to baseline levels while IFNG mRNA stayed higher than baseline. IL17A was undetectable at baseline and remained undetectable at each time point.

Figure 2.

 Monitoring IFNG and IL4 mRNA expression during and after rituximab therapy: change in IFNG and IL4 mRNA expression in the blood during rituximab therapy. Data are shown as the means of triplicate ± standard error (SE). The results were analysed using Graphpad Prism 5 software. Unpaired student’s t-test was performed and the differences were considered significant when < 0·05.

In our patient, the first dose of rituximab depleted circulating B cells and eliminated the FVIII inhibitor by Bethesda assay. However, despite continued B-cell depletion, the inhibitor slowly reappeared over time, suggesting that re-emergence of the inhibitor may have been due to residual B cells maturing into plasma cells that do not express CD20 (Lanzavecchia & Sallusto, 2009).

B cells are known to control IFN-γ production by T-cells through a dense network of anti- inflammatory cytokines, such as IL-10 (Moore et al, 2001; Lund, 2008). The present report demonstrated that increasing levels of IFNG mRNA expression correlated with initial B-cell depletion, suggesting that B-cells may suppress T-cell production of IFN-γ. This is intriguing and, to our knowledge, similar findings have not yet been clearly reported elsewhere, although the mechanism by which B-cell removal is associated with increased expression of IFNG is not known. However, it has been reported that rituximab-induced B-cell depletion may exacerbate some T-cell dependent immune mediated diseases, such as psoriasis (Dass et al, 2007) and ulcerative colitis (Goetz et al, 2007). This suggests that T cell activity changes in the absence of B cells, which may explain why certain T-cell mediated diseases, such as ulcerative colitis and psoriasis, may be exacerbated after rituximab therapy. The balance between IFN-γ and IL-4 has been shown to play a crucial role in the control of a variety of latent infections and auto immune conditions (Lucey et al, 1996; Livingston et al, 1999).

Our findings strengthen prior thoughts that the effects of rituximab go beyond depleting B cells, but appear to affect Th1 and Th2 balances that may be at work in the eradication of inhibitors. Further investigation into the behaviour of these cell subsets will be helpful in the better understanding of inhibitor elimination.


This work was supported by the Louisiana Comprehensive Hemophilia Care Center.